Machine learning has been a subject undergoing intense study across many different industries and academic research areas. Companies and researchers have taken full advantages of various machine learning approaches to solve their problems; however, vast understanding and study of the field is required for developers to fully harvest the potential of different machine learning models and to achieve efficient results. Therefore, this thesis begins by comparing auto ml with other hyper-parameter optimization techniques. auto ml is a fully autonomous framework that lessens the knowledge prerequisite to accomplish complicated machine learning tasks. The auto ml framework automatically selects the best features from a given data set and chooses the best model to fit and predict the data. Through multiple tests, auto ml outperforms MLP and other similar frameworks in various datasets using small amount of processing time. The thesis then proposes and implements a stacking ensemble technique in order to build protection against over-fitting for small datasets into the auto ml framework. Stacking is a technique used to combine a collection of Machine Learning models’ predictions to arrive at a final prediction. The stacked auto ml ensemble results are more stable and consistent than the original framework; across different training sizes of all analyzed small datasets. / Master of Science

There once was a dream that data-driven models would replace their theory-guided counterparts. We have awoken from this dream. We now know that data cannot replace theory. Data-driven models still have their advantages, mainly in computational efficiency but also providing us with some special sauce that is unreachable by our current theories. This dissertation aims to provide a way in which both the accuracy of theory-guided models, and the computational efficiency of data-driven models can be combined. This combination of theory-guided and data-driven allows us to combine ideas from a much broader set of disciplines, and can help pave the way for robust and fast methods. / Doctor of Philosophy / As an illustrative example take the problem of predicting the weather. Typically a supercomputer will run a model several times to generate predictions few days into the future. Sensors such as those on satellites will then pick up observations about a few points on the globe, that are not representative of the whole atmosphere. These observations are combined, ``assimilated'' with the computer model predictions to create a better representation of our current understanding of the state of the earth. This predict-assimilate cycle is repeated every day, and is called (sequential) data assimilation. The prediction step traditional was performed by a computer model that was based on rigorous mathematics. With the advent of big-data, many have wondered if models based purely on data would take over. This has not happened. This thesis is concerned with taking traditional mathematical models and running them alongside data-driven models in the prediction step, then building a theory in which both can be used in data assimilation at the same time in order to not have a drop in accuracy and have a decrease in computational cost.

Background: Epidemiological studies suggest an increased incidence of osteoarthritis among workers in occupations requiring squat-lifting such as in construction, mining and farming. Squat-lifting postures can induce heavy mechanical loads on the joint, causing the articulating surfaces to deform. This can result in changes of vibration characteristics of the joint surfaces. Differences in the vibration characteristics of normal and pathological joints have been established and used in the past for classifying severity of disease. The purpose of this study was to examine the influence of cumulative mechanical load on the vibration properties of the knee joint and to gain an understanding of how these properties may relate to an increase in cumulative load placed on the joint. Methods: In this study, cumulative load was measured as the resultant knee joint torque during squat lifting, while a piezoelectric accelerometer was used to capture vibration signals from points on the knee during flexion and extension. Twelve university students were recruited for a repeated measures study. Each participant attended one session where they had to perform a series of six squat-lifting tasks on a force platform. Motion capture equipment was used to obtain kinematic data. The cumulative 3-D moment on the joint was calculated using inverse dynamics. Results: A visual inspection of an ensemble average constructed for the frequency spectrum of all participants revealed that differences may exist in the 750 Hz - 2000 Hz bandwidth for vibrations coming from the patella during flexion. Further statistical analysis by a t-test and ANOVA showed a decrease in the RMS power of the signal captured in this bandwidth before and after mechanical load was induced by squat lifting. A linear regression analysis indicated a significant correlation between cumulative 3-D moment on the knee joint and the median frequency of vibration signals from the patella during flexion in the 1000 Hz - 2500 Hz range. Conclusions: Overall, the results of this study indicate the possibility of a relationship between mechanical exposure on the knee joint and its vibration properties during joint movement. Despite the small sample size, a declining trend was observed in the normalized RMS power of signals with increase in loading. However, the quantitative nature of this relationship is not clear and the current study points towards a non-linear relationship between joint exposure and knee vibrations. Future studies must investigate this possibility using direct measures of joint loading, cartilage deformation and their relation to joint vibrations. / Master of Science

Denna kvalitativa studies syfte är att undersöka hur sångare inkluderas och exkluderas som en del av en ensemble, samt vilka didaktiska val läraren gör som bidrar till inkludering eller exkludering av sångaren. Forskningsfrågorna som ligger till grund för det självständiga arbetet var: På vilka sätt inkluderas sångaren som en del av en ensemble? På vilka sätt exkluderas sångaren som en del av en ensemblen? Vilka didaktiska val använder sig lärare av för inkludering och exkludering av sångare i ensemble? Resultatet genererades genom observationer av tre olika ensemblelärare på gymnasiets estetiska linje, vilket sedan analyserades utifrån teorin Community of practice, som bygger på att lärande sker i samspelet mellan människor inom given praktik. Resultatet visar att inkludering av sångaren uppstår till följd av en rad olika didaktiska verktyg, både genom att läraren låter instrumentspecifika instruktioner ta plats under den lärarledda tiden, att läraren medvetet inkluderar sångaren i ensemblens gemensamma förhandlande och genom att använda ett “språk” som anpassas till allas kunskapsnivå. Språkbruk och ordval är också sätt som läraren exkluderar sångaren på, dels genom att inte ge kunskap kring notbilden men också då läraren väljer att förenkla musiktermer. Det didaktiska verktyget att dela upp sångare och instrumentalister leder också till exkludering, vilket försätter sångaren i en position av icke fullt deltagande i ensemblen. Resultatet visar på att en större medvetenhet hos ensemblelärare, gällande vokalistpositionen och vad denna position innefattar, är nödvändig för att en så jämlik ensembleundervisning som möjligt ska kunna uppnås.

Los modelos de aprendizaje automático están siendo ampliamente utilizados en múltiples áreas de la vida humana. Tradicionalmente, se han aplicado en reconocimiento de voz, detección de rostros, clasificación de imágenes, sistemas de recomendación, etc. Con la reciente revolución de los modelos generativos, la popularidad de los chatbots conversacionales se ha disparado. Esto ha dado lugar a que los modelos de aprendizaje automático se utilicen cada vez más para abordar tareas para las que no estaban específicamente capacitados. El prompt engineering ha permitido que personas no expertas en aprendizaje automático (que comúnmente tampoco están familiarizadas con los problemas subyacentes al uso de modelos de aprendizaje automático para hacer predicciones) automaticen ciertas tareas. La incorporación de algoritmos de aprendizaje automático en tareas de toma de decisiones de alto riesgo ha levantado algunas alertas en la comunidad científica. Las tareas de toma de decisiones de alto riesgo denotan aquellas tareas que pueden tener un gran impacto en las vidas de las personas sobre quienes se toman las decisiones. Por ejemplo, se han utilizado modelos para decidir si una persona es contratada o no, si se le concede un préstamo, si se acepta una solicitud de cobertura ampliada de seguridad sanitaria y para predecir la probabilidad de reincidencia en un delito. Estudios han demostrado que la automatización inconsciente de este tipo de tareas contiene sesgos, lo cual provoca que decisiones injustas sean tomadas sobre determinados grupos de población. El peligro fundamental de ignorar este problema es que los métodos de aprendizaje automático podrían no sólo reflejar los sesgos presentes en nuestra sociedad, sino que también amplificarlos. Esta tesis presenta el diseño y validación de una tecnología para asistir la automatización justa de problemas de clasificación. En esencia, la propuesta se basa en diseñar una tecnología que saque provecho de las soluciones intermedias generadas durante la resolución de problemas de clasificación mediante el uso de herramientas de Auto-ML, en particular, AutoGOAL, con el propósito de crear clasificadores imparciales y justos. Cuatro componentes fundamentales conforman la propuesta: (I) una componente de Auto-ML, encargada de generar colecciones de modelos con hipótesis diferentes entre sí, cada uno debiendo ser capaz de transformar los datos de entrada a los de salida; (II) una componente de ensemble, responsable de combinar múltiples soluciones para producir una más robusta, exacta y justa, según criterios definidos por funciones de pérdida; (III) una componente de cuantificación de sesgos, encargada de medir la calidad de un modelo según varias definiciones de sesgo y equidad; y (IV) una componente de anotación automática de atributos protegidos, para permitir estimar los sesgos en situaciones donde normalmente no sería posible. Los resultados discutidos en este documento prueban la capacidad de la tecnología propuesta para automatizar tareas de clasificación a la par que se controla su equidad. Además, la experimentación muestra la competitividad de la propuesta respecto a otras alternativas para controlar los sesgos en el modelo de aprendizaje automático y colecciones de datos. La propuesta tiene la ventaja adicional de requerir poco o ningún conocimiento sobre arquitecturas de aprendizaje automático dada las componentes de Auto-ML que incorpora. Todo esto lo convierte en una tecnología valiosa y conveniente de explotar. Más allá de los resultados obtenidos sobre la línea principal de investigación, se obtuvieron algunos resultados secundarios de gran valor. Por un lado, se pudo demostrar que ciertas componentes de la tecnología pueden utilizarse para extender, de forma más robusta, corpus de texto con entidades y relaciones anotadas, asumiendo que se cuente con varias versiones de anotación que se quieran combinar; esta condición suele ser viable dada la existencia de competencias internacionales donde varios participantes compiten por generar la mejor solución. Muy ligado a esto, otro resultado secundario es la generación del eHealth-KD 2019 ensembled corpus a partir de ensamblar las soluciones participantes de la competencia eHealth-KD 2019. Otro resultado secundario se tiene en la componente de anotación automática de atributos protegidos, la cual puede funcionar independiente del flujo end-to-end de automatización justa, y por tanto puede utilizarse como fase de preprocesamiento para otros sistemas de cuantificación o mitigación de sesgos. En ese sentido, se publicó también el corpus Reviews' Gender} diseñado para auxiliar el entrenamiento y evaluación de modelos de anotación de atributos protegidos con el fin de usar sus predicciones como estimadores de equidad. Todos los recursos desarrollados en esta investigación están a disposición pública de la comunidad científica. / Esta investigación ha sido desarrollada de forma conjunta en la Universidad de Alicante (España) y la Universidad de La Habana (Cuba), entre enero de 2020 y septiembre de 2023, en sucesivas estancias de investigación co-financiadas por ambas instituciones. La Universidad de Alicante, el Departamento de Lenguajes y Sistemas Informáticos ha soportado esta investigación a través de los proyectos TRIVIAL (PID2021-122263OB-C22), CORTEX (PID2021-123956OB-I00), CLEARTEXT (TED2021-130707B-I00), SOCIALTRUST (PDC2022-133146-C22), NL4DISMIS (CIPROM/2021/21) y VIVES (2022-TL22-00215334). La Universidad de La Habana, la Facultad de Matemática y Computación y el Departamento de Inteligencia Artificial y Sistemas Computacionales han soportado esta investigación.

This study addresses the ways and means a conductor may approach two major twentieth century works written specifically for percussion ensemble. Performance techniques and decisions on aesthetics made by the conductor in dealing with such items as timbre, balance, pitch levels, and pitch relationships are also considered.

Syftet med föreliggande studie är att utifrån ett genusperspektiv studera det sociala och musikaliska samspelet i ensembleundervisning på högstadiet och gymnasiet. Studien utgår från ett genusperspektiv där genus ses som en social konstruktion. Datamaterialet består av videoobservationer av tre musiklektioner där ensemblespel stod på schemat. Resultatet visar att eleverna aktivt återskapade könsstereotypa instrumentfördelningar och på så vis befäste den heteronormativa hierarkin. Resultatet visade dessutom ett mer dominant platstagande hos vissa killar i grupperna och ett generellt sett mer understödjande beteendemönster hos tjejerna i grupperna. Diskussionen antyder att ramarna för handlingsutrymme är snävare för tjejerna än för killarna (bland annat i och med kopplingen till de informella läroprocesserna i undervisningen) och erbjuder genuskontraktets regelsystem som möjlig förklaringsmodell för tjejers och killars olika platstagande. / The purpose of this study is to examine, from a gender point of view, the social and musical interaction in ensemble groups during ensemble lessons at secondary levels. The study is based on a gender perspective where gender is seen as a social construction. The data consists of video observations of three separate music lessons where ensemble playing was on the curriculum. The result showed the pupils actively recreating gender stereotypical instrument allocations and, in doing so, consolidating the heteronormative hierarchy. Furthermore, the result showed a more dominant behaviour amongst certain of the boys in the groups, and amongst the girls a predominantly more supporting behavioural pattern. The discussion suggests the space for action for girls being narrower than that for the boys, and presents the set of rules accompanying the gender contract as possible key for understanding girls' and boys' different place-taking.

We solve a random matrix ensemble called the chiral Ginibre orthogonal ensemble, or chGinOE. This non-Hermitian ensemble has applications to modelling particular low-energy limits of two-colour quantum chromo-dynamics (QCD). In particular, the matrices model the Dirac operator for quarks in the presence of a gluon gauge field of fixed topology, with an arbitrary number of flavours of virtual quarks and a non-zero quark chemical potential. We derive the joint probability density function (JPDF) of eigenvalues for this ensemble for finite matrix size N, which we then write in a factorised form. We then present two different methods for determining the correlation functions, resulting in compact expressions involving Pfaffians containing the associated kernel. We determine the microscopic large-N limits at strong and weak non-Hermiticity (required for physical applications) for both the real and complex eigenvalue densities. Various other properties of the ensemble are also investigated, including the skew-orthogonal polynomials and the fraction of eigenvalues that are real. A number of the techniques that we develop have more general applicability within random matrix theory, some of which we also explore in this thesis.

O crescimento acelerado da internet proporcionou uma quantidade grande de informações acessíveis aos usuários. Ainda que tal quantidade possua algumas vantagens, os usuários que possuem pouca ou nenhuma experiência para escolher uma alternativa dentre as várias apresentadas terão dificuldades em encontrar informações (ou itens, considerando o escopo deste trabalho) úteis e que atendam às suas necessidades. Devido a esse contexto, os sistemas de recomendação foram desenvolvidos para auxiliar os usuários a encontrar itens relevantes e personalizados. Tais sistemas são divididos em diversas arquiteturas. Como exemplo estão as arquiteturas baseadas em: conteúdo, filtro colaborativo e conhecimento. Para este trabalho, a primeira arquitetura foi explorada. A arquitetura baseada em conteúdo recomenda itens ao usuário com base na similaridade desses aos itens que o usuário mostrou interesse no passado. Por consequência, essa arquitetura possui a limitação de, geralmente, realizar recomendações com baixa serendipidade, uma vez que os itens recomendados tendem a ser semelhantes àqueles observados pelo o usuário e, portanto, não apresentam novidade ou surpresa. Diante desta limitação, o aspecto de serendipidade tem destaque nas discussões apresentadas neste trabalho. Assim, o objetivo deste trabalho é minimizar o problema da baixa serendipidade das recomendações por meio da utilização da análise de similaridades parciais implementada usando ensemble de agrupamentos. Para alcançar este objetivo, estratégias de recomendação baseadas em conteúdo implementadas usando agrupamento e ensemble de agrupamento foram propostas e avaliadas neste trabalho. A avaliação contou com análises qualitativas sobre as recomendações produzidas e com um estudo com usuários. Nesse estudo, quatro estratégias de recomendação de notícias foram avaliadas, incluindo as duas propostas neste trabalhos, uma estratégia baseada em recomendação aleatória, e uma estratégia baseada em coagrupamento. As avaliações consideraram aspectos de relevância, surpresa e serendipidade de recomendações. Esse último aspecto é descrito como itens que apresentam tanto surpresa quanto relevância ao usuário. Os resultados de ambas análises mostraram a viabilidade da utilização de agrupamento como base de recomendação, uma vez que o ensemble de agrupamentos obteve resultados satisfatórios em todos os aspectos, principalmente em surpresa, enquanto a estratégia baseada em agrupamento simples obteve os melhores resultados em relevância e serendipidade / The accelerated growth of the internet has provided a large amount of information accessible to users. Although this amount of information has some advantages, users who have little or no experience in choosing one of several alternatives will find it difficulty to find useful information (or items, considering the scope of this work) that meets their needs. Due to this context, recommender systems have been developed to help users find relevant and personalized items. Such systems are divided into several architectures as content-based, collaborative filtering and knowledge-based. The first architecture was explored in this work. The content-based architecture recommends items to the user based on their similarity to items that the user has shown interest in the past. Consequently, this architecture has the limitation of generally making recommendations with low serendipity, since the recommended items tend to be similar to those observed by the user and, therefore, do not present novelty or surprise. Given this limitation, the aspect of serendipity is highlighted in the discussions presented in this work. Thus, the objective of this work is to minimize the problem of the low serendipity of the recommendations through the use of the partial similarity analysis implemented using cluster ensemble. To achieve this goal, content-based recommendation strategies implemented using clustering and cluster ensemble were proposed and evaluated. The evaluation involved qualitative analysis of the recommendations and a study with users. In such a study, four news recommendation strategies were evaluated including the two strategies proposed in this work, a strategy based on random recommendation, and a strategy based on co-clustering. The evaluations considered aspects of relevance, surprise and serendipity of recommendations. This last aspect is described as items that present both surprise and relevance to the user. The results of both analyzes showed the feasibility of using clustering as the basis of recommendation, since cluster ensemble had satisfactory results in all aspects, mainly in surprise, whereas the simple clustering-based strategy obtained the best results in relevance and serendipity

O Filtro de Kalman Ensemble (EnKF) é um algoritmo de Monte Carlo sequencial para inferência em modelos de espaço de estados lineares e não lineares. Este filtro combinado com alguns outros métodos propaga a distribuição a posteriori conjunta dos estados e dos parâmetros ao longo do tempo. Existem poucos trabalhos que consideram o problema da estimação simultânea dos estados e parâmetros, e os métodos existentes possuem limitações. Nesta dissertação analisamos a eficiência desses métodos por meio de estudos de simulação em modelos de espaço de estados lineares e não lineares. O problema de estimação não linear aqui tratado refere-se ao modelo de produção excedente logístico, para o qual o EnKF pode ser considerado uma possível alternativa aos algoritmos MCMC. Os resultados da simulação revelam que a acurácia das estimativas aumenta quando a série temporal cresce, mas alguns parâmetros apresentam problemas na estimação. / The Ensemble Kalman Filter (EnKF) is a sequential Monte Carlo algorithm for inference in linear and nonlinear state-space models. This filter combined with some other methods propagates the joint posterior distribution of states and parameters over time. There are fewer papers that consider the problem of simultaneous state-parameter estimation and existing methods have limitations. The purpose of this dissertation is to analyze the efficiency of these methods by means of simulation studies in linear and nonlinear state-space models. The nonlinear estimation problem addressed here refers to the logistic surplus-production model, for which the EnKF can be considered as a possible alternative to MCMC algorithms. The simulation results reveal that the accuracy of the estimates increases when the time series grows, but some parameters present problems in the estimation.